Andrea Griffiths
Senior Developer Advocate
Sarasota, Florida, United States
Actions
Andrea Griffiths is a Senior Developer Advocate at GitHub. She hosts The GitHub Podcast, Open Source Friday, and GitHub Checkout, and writes Main Branch, a weekly newsletter on developer fundamentals. She learned to code through freeCodeCamp in 2016 while raising her kids, made a career pivot in her mid-30s, and has been building developer tools ever since. Based on Florida's west coast, she creates content in English and Spanish and has deep ties to the Latin American developer community.
Area of Expertise
Tu Repositorio Sabe Más de Tu Equipo Que Tu Equipo
He entrado a equipos donde pasé meses siendo invisible. No porque la gente fuera mala gente, sino porque todo el conocimiento estaba enterrado en 50.000 commits que nunca había visto. ¿Quién le sabe al flow de autenticación? ¿A quién le gusta debuggear race conditions? No tenía idea. Estaba adivinando.
Entonces construí Team X-Ray — una extensión de VS Code que analiza personas, no código: cómo se comunican, con quién colaboran, en qué son buenos de verdad, qué problemas disfrutan resolver. Empecé conectando GitHub Copilot con los datos del repositorio usando Model Context Protocol (MCP), pero la arquitectura evolucionó — hoy usa el Copilot SDK directamente con herramientas custom que le dan al agente acceso nativo a los datos de tu equipo. Sin Docker, sin servidor intermedio.
Y luego pasó algo que no esperaba: los agentes de IA empezaron a hacer commits. De repente, mi herramienta me estaba mintiendo — atribuyéndole expertise a bots, inflando contribuciones que ningún humano escribió. Tuve que evolucionar la herramienta para separar lo humano de lo automático, y en el camino descubrí algo más interesante: en la era de agentes, la expertise ya no es solo quién escribe el código — es quién lo entiende, quién lo dirige, y quién hace brillar a los demás.
En esta charla te voy a mostrar:
• La evolución técnica: de MCP a Copilot SDK con herramientas custom
• Cómo construí Team X-Ray y qué me enseñó sobre dinámicas de equipo reales
• El problema de los agentes: cómo detectar commits de bots y separar expertise humana de la automatizada
• Ejemplos reales de IA descubriendo mentores, silos de conocimiento y dinámicas ocultas
Como dijo Moira Rose: "Cuando uno de nosotros brilla, todos brillamos." No estoy tratando de reemplazar la conexión humana. Estoy tratando de acelerarla — incluso cuando los agentes ya son parte del equipo.
When AI Agents Go Rogue: Securing the New Attack Surface
AI agents are going from helpful assistants to autonomous teammates with real system access - but what happens when they go rogue? As AI agents gain access to repositories, infrastructure, and sensitive data through the Model Context Protocol (MCP), we're seeing new attack vectors that traditional security tools can't handle: AI worms that self-replicate through prompts, malicious MCP servers, and automated credential theft at unprecedented scale. GitHub became the first platform to build secret scanning directly into AI tool calls. We'll demonstrate live attacks, show real-time protection in action, and give you practical strategies to secure your AI workflows today - before your AI agents accidentally become your biggest security vulnerability.
Welcome to OctogatosConf 2022 Keynote by @mariorod
We are grateful to have you here, for a day of learning to network and shine!
The AI Teammate Is Here — Now What?
We're not in the "AI as tooling" phase anymore. Your developers are working with agents that co-author PRs, copilots that suggest architecture decisions, and systems that know your codebase patterns better than your junior devs do. As engineering leaders, this changes everything. You're not just managing people - you're managing workflows where humans and AI collaborate, delegate, and sometimes disagree on the best approach.
At GitHub, I get to work with teams navigating this shift daily. The successful ones aren't just adopting AI tools faster. They're rethinking how work gets done, how decisions get made, and how trust gets built when your "team" includes non-human contributors.
What we'll cover:
- What "agentic engineering" actually looks like in practice (spoiler: it's messier than the demos)
- How to lead developers who are learning to work with AI, not just use it
- Building trust in AI systems without losing your humans in the process
You'll walk away with:
- Tips for hybrid human-AI workflows
- Practical strategies for scaling AI adoption without team burnout
- Tools for maintaining code quality and engineering culture as AI becomes a bigger contributor
This is happening in teams just like yours right now. Come ready for honest conversations about what works, what doesn't, and how to lead through the messiness.
Preparación y Examen de Práctica para la Certificación Fundamentos de GitHub
Esta sesión intensiva está diseñada para prepararte para la certificación Fundamentos de GitHub. A través de una combinación de teoría, ejercicios prácticos y un examen de práctica, cubriremos aspectos cruciales del examen. Exploraremos en profundidad cómo navegar eficientemente por GitHub, maximizar el uso de sus herramientas colaborativas y gestionar proyectos con maestría. Como incentivo adicional, todos los asistentes recibirán un voucher gratuito para realizar el examen de certificación oficial de Fundamentos de GitHub. Esta es una oportunidad única para no solo prepararte, sino también para obtener tu certificación sin costo adicional.
OSS to level up your career, reputation or to get your first job.
Regardless of where you are in your developer journey chances are one or more of your projects are using OSS, there is a space in the OSS ecosystem for contributions of all types and skill levels.
Contributing to open source projects is an effective way to build up your reputation, level up your skills or get your first job. In this talk, I'll reveal strategies for optimizing your OSS contributions, ensuring they are closely aligned with your career objectives. Even if you haven't begun contributing yet, you'll walk away with a clear contribution plan.
Dominando GitHub: Estrategias y Preparación para la Certificación de Fundamentos
Dominando GitHub: Estrategias y Preparación para la Certificación de Fundamentos
Participa en nuestro taller especializado diseñado para desarrolladores, por desarrolladores, con el objetivo de dominar GitHub a un nivel fundamental. Este taller técnico está estructurado para proporcionar a los ingenieros y profesionales del desarrollo de software las competencias esenciales para manejar eficazmente proyectos en GitHub, optimizar la colaboración y garantizar la seguridad y administración efectiva de repositorios. Será impartido por Andrea Griffiths, Senior PM, y Javier Perez, Senior Engineering Manager, ambos con una vasta experiencia y trayectoria en GitHub.
Perfil del Público:
Ideal para desarrolladores, ingenieros de software y profesionales técnicos que utilizan o planean utilizar GitHub para la gestión de versiones y colaboración. Este taller es crucial para aquellos que buscan certificar su habilidad en GitHub y aplicar prácticas avanzadas en sus flujos de trabajo de desarrollo.
Objetivos del Taller
1. Git y GitHub: Profundización en sistemas de control de versiones distribuidos, con énfasis en la integración y diferenciación entre Git y GitHub. Análisis de repositorios, commits, ramas y flujo de trabajo de GitHub.
2. Características de Colaboración: Técnicas avanzadas para la gestión de pull requests e issues, estrategias para el manejo eficaz de discusiones y colaboraciones en proyectos de software.
3. Desarrollo Moderno: Exploración de GitHub Actions para automatización CI/CD, introducción a GitHub Copilot para asistencia de código y optimización de entornos de desarrollo con GitHub Codespaces.
4. Gestión de Proyectos: Aplicación de GitHub Projects para la planificación y seguimiento de tareas, uso de etiquetas y hitos para la organización de proyectos complejos.
5. Privacidad, Seguridad y Administración: Mejores prácticas en la configuración de seguridad de repositorios, autenticación de dos factores (2FA), y administración de permisos y políticas de colaboración.
6. Contribución a la Comunidad: Comprensión del impacto y la gestión de proyectos open source, con enfoque en la contribución efectiva y el aprovechamiento de la comunidad de GitHub para la innovación colaborativa.
Contenido y Recursos:
Acceso a material diseñado en colaboración con GitHub, que abarca desde fundamentos técnicos hasta escenarios de uso avanzado, con rutas de aprendizaje específicas de Microsoft Learn y LinkedIn Learning. Los participantes se enfrentarán a ejercicios prácticos y casos de estudio para afianzar conocimientos técnicos y prepararse para el examen de certificación.
Beneficios del Taller:
- Capacitación técnica profunda por expertos en la materia, Andrea Griffiths y Javier Perez de GitHub.
- Ejercicios prácticos centrados en resolver desafíos de ingeniería real.
- Acceso a una comunidad de profesionales técnicos y oportunidades de networking.
- Vale por 200 USD válido para la certificación de Fundamentos de GitHub, agregando un valor distintivo a tu perfil profesional (limitado a las primeras 100 personas que finalicen el itinerario de Fundamentos).
Asegura tu participación en este taller técnico avanzado, diseñado para elevar tus habilidades en GitHub al siguiente nivel y obtener la certificación que validará tu expertise ante empleadores y colegas.
Building Mobile Apps with GitHub Copilot + MCP
This hands-on workshop walks participants through building a mobile application using GitHub Copilot, the Model Context Protocol (MCP), and the Spotify API. We’ll start with an overview of MCP and demonstrate GitHub’s MCP server in action. Then we’ll build a real mobile app, powered by AI and GitHub metadata. Participants will learn how to draft a Product Requirements Document (PRD) with GitHub Copilot, generate UI with Claude, and use Copilot Agent Mode to ship a personalized playlist experience. They will also learn how to get better results when building with AI tools by implementing prompt files and custom instructions.
Building Mobile Apps with GitHub Copilot + MCP
This hands-on workshop walks participants through building a mobile application using GitHub Copilot and the GitHub Model Context Protocol (MCP). We’ll start with an overview of MCP and demonstrate GitHub’s MCP server in action. Then we’ll build a real mobile app, powered by AI and GitHub metadata. Participants will learn how to draft a Product Requirements Document (PRD) with GitHub Copilot, generate UI with Claude, and use Copilot Agent Mode to ship a personalized playlist experience. They will also learn how to get better results when building with AI tools by implementing prompt files and custom instructions.
Required Setup:
- GitHub Account: Create free account at https://github.com if you don't have one
- GitHub Copilot access:
- Free: 50 chats/month + 2000 completions at https://github.com/settings/copilot
- Students: Free Copilot Pro with GitHub Student Developer Pack at https://education.github.com
- Pro: $10/month for unlimited access at https://github.com/settings/copilot
Workshop Environment (Choose One):
Option 1: GitHub Codespace (Recommended)
- Click the workshop Codespace link (provided by instructor)
- Wait 3-5 minutes for container to build (includes Flutter, extensions, and dependencies)
- Everything pre-configured and ready to use
Option 2: Local Development (Alternative)
- Flutter 3.7+: installed and working (flutter doctor passing) - Get it at https://flutter.dev/docs/get-started/install
- VS Code: with Flutter extension OR IntelliJ IDEA/Android Studio with Flutter plugin
- Device/emulator: ready for testing (iOS Simulator, Android Emulator, or Chrome browser)
Required API Accounts (Free Setup):
- GitHub Personal Access Token: Create at https://github.com/settings/tokens with scopes: repo, user, read:org
- Spotify Developer Account: Create free account and app at https://developer.spotify.com/dashboard
- App name: "GitVision Workshop"
- Redirect URI: gitvision://callback
- Copy Client ID and Client Secret
Programming Knowledge Required:
- Basic Dart/Flutter understanding (widgets, async/await, JSON parsing)
- REST API concepts (HTTP requests, authentication, error handling)
- AI development experience with prompt engineering and AI-assisted coding tools
Duration: 120 minutes
Format: AI-assisted live coding in GitHub Codespaces
Skill Level: Intermediate developers familiar with AI development tools
GitHub’s Team X-Ray: Your Repository Knows More About Your Team Than Your Team Does
I've joined teams where I spent months being invisible.
Not because people were bad. Just because all the knowledge was buried in 50,000 commits I'd never seen. Who owns the auth flow? Who actually likes debugging race conditions? I had no idea. I was guessing, and it showed.
So I built Team X-Ray. It's a VS Code extension that analyzes people, not code. How they communicate. Who they collaborate with. What they're genuinely good at. What kinds of problems they actually enjoy solving.
It started as a GitHub Copilot integration connected to repository data through MCP. But the architecture evolved. Today it talks directly to the Copilot SDK with custom tools that give the agent native access to your team's data. No Docker. No middleware server.
Then something unexpected happened: AI agents started making commits.
Suddenly my tool was lying to me. It was attributing expertise to bots. Inflating contributions that no human wrote. I had to teach it to separate human work from automated work, and in doing that I stumbled onto something more interesting.
In the age of agents, expertise isn't just who writes the code. It's who understands it. Who directs it. Who makes everyone else better.
In this talk I'll walk you through:
- The technical evolution: from MCP to Copilot SDK with custom tools
- How I built Team X-Ray and what it taught me about real team dynamics
- The agent problem: how to detect bot commits and separate human expertise from automation
- Real examples of AI surfacing mentors, knowledge silos, and hidden team dynamics
As Moira Rose said: "When one of us shines, we all shine."
I'm not trying to replace human connection. I'm trying to accelerate it. Even when agents are already part of the team.
Nebraska.Code() 2026 Sessionize Event Upcoming
The Commit Your Code Conference 2025! Sessionize Event
swampUP 2025 Sessionize Event
Andrea Griffiths
Senior Developer Advocate
Sarasota, Florida, United States
Actions
Please note that Sessionize is not responsible for the accuracy or validity of the data provided by speakers. If you suspect this profile to be fake or spam, please let us know.
Jump to top